Research

Learning to Scale Multilingual Representations for Vision-Language Tasks

ECCV

Authors

Published on

08/28/2020

Categories

ECCV

Current multilingual vision-language models either require a large number of additional parameters for each supported language, or suffer performance degradation as languages are added. In this paper, we propose a Scalable Multilingual Aligned Language Representation (SMALR) that supports many languages with few model parameters without sacrificing downstream task performance. SMALR learns a fixed size language-agnostic representation for most words in a multilingual vocabulary, keeping language-specific features for just a few. We use a masked cross-language modeling loss to align features with context from other languages. Additionally, we propose a cross-lingual consistency module that ensures predictions made for a query and its machine translation are comparable. The effectiveness of SMALR is demonstrated with ten diverse languages, over twice the number supported in visionlanguage tasks to date. We evaluate on multilingual image-sentence retrieval and outperform prior work by 3-4% with less than 1/5th the training parameters compared to other word embedding methods.

This work was published in ECCV 2020.

Please cite our work using the BibTeX below.

@inproceedings{burns2020eccv, 
title={Learning to Scale Multilingual Representations for Vision-Language Tasks},
author={Andrea Burns and Donghyun Kim and Derry Wijaya and Kate Saenko and Bryan A. Plummer}, 
booktitle={The European Conference on Computer Vision (ECCV)}, 
year={2020} 
}  
Close Modal